REvIEWS - Alzheimer Autrement

reviews
Changing perspectives regarding
late-life dementia
Majid Fotuhi, Vladimir Hachinski and Peter J. Whitehouse
Abstract | Individuals over 80 years of age represent the most rapidly growing segment of the population, and
late-life dementia has become a major public health concern worldwide. Development of effective preventive
and treatment strategies for late-life dementia relies on a deep understanding of all the processes involved.
In the centuries since the Greek philosopher Pythagoras described the inevitable loss of higher cognitive
functions with advanced age, various theories regarding the potential culprits have dominated the field, ranging
from demonic possession, through ‘hardening of blood vessels’, to Alzheimer disease (AD). Recent studies
suggest that atrophy in the cortex and hippocampus—now considered to be the best determinant of cognitive
decline with aging—results from a combination of AD pathology, inflammation, Lewy bodies, and vascular
lesions. A specific constellation of genetic and environmental factors (including apolipoprotein E genotype,
obesity, diabetes, hypertension, head trauma, systemic illnesses, and obstructive sleep apnea) contributes
to late-life brain atrophy and dementia in each individual. Only a small percentage of people beyond the age
of 80 years have ‘pure AD’ or ‘pure vascular dementia’. These concepts, formulated as the dynamic polygon
hypothesis, have major implications for clinical trials, as any given drug might not be ideal for all elderly people
with dementia.
s
w
e
i
v
D
e
E
R
T
e
r
C
u
E
t
R
a
N OR OF
C
O
N
R
U
P
Fotuhi, M. et al. Nat. Rev. Neurol. advance online publication XX Month 2009; doi:10.1038/nrneurol.2009.175
Introduction
Memory loss, dementia, and Alzheimer disease (AD)
are major public health concerns worldwide. In recent
years, AD has become almost synonymous with latelife dementia. 100 years ago, however, senile dementia
(mostly attributed to ‘hardening of the blood vessels’)
was considered to be the dominant etiology for cognitive impairment in elderly individuals over the age
of 80 years. 1,000 years ago, demonic possession was
blamed for the same set of dementia symptoms. Clearly,
scientists in each era have tried to untangle the complex
etiology of late-life dementia, and still no specific effective remedy has emerged.
In this critical Review of the dementia literature, we
trace the development of various dominant concepts and
theories and outline a set of key discoveries that have
brought us to our current state of knowledge in this field.
We focus on the most recent studies, which suggest that
cognitive impairment among the oldest old results from
a dynamic complex of genetic and environmental factors.
We discuss the implications of these developments for
the design of future clinical trials and summarize some
of the key questions that we must answer in the coming
years. For example, is late-life dementia an extension of
AD pathology, or is it qualitatively and quantitatively
different from early-onset AD? Also, what are the best
and most specific biomarkers and imaging techniques
to detect presymptomatic cognitive impairment and to
Competing interests
The authors declare no competing interests.
nature reviews | neurology monitor the rate of clinical progression in elderly indivi­
duals with dementia?
Evolution of concepts
Greco-Roman period to 1907
Cognitive decline with aging was described by Western
philosophers and clinicians as early as the 7th century
BC.1 The Greek philosopher Pythagoras observed that
the pattern of development of new skills early in life
reverses toward the end of life. ‘Normal’ regression of
mental faculties, according to Pythagoras, would begin
in one’s 60s and, by one’s 80s, would lead to the “imbecility of infancy”. These concepts persisted until the Early
Renaissance period, when patients with dementia were
treated as witches. In the 18th century, ‘senile dementia’ was recognized as a distinct condition from normal
aging, and patients with this condition were shown to
have smaller brains, on average, than their cognitively
healthy counterparts.2 In the 1890s, Alois Alzheimer
and Otto Binswanger extensively described and emphasized the critical roles of atherosclerosis and stroke in the
development of brain atrophy and senile dementia.2
1907–1997
Alois Alzheimer’s findings of plaques and tangles during
the autopsy evaluation of a young patient with progressive confusion and hallucination were published in 1907
as a case report entitled “About a peculiar disease of the
cerebral cortex”. In 1910, Emil Kraepelin, in part for
political purposes in the context of rivalry between two
Center for Memory and
Brain Health, The
Sandra and Malcolm
Berman Brain & Spine
Institute, Baltimore,
MD, USA (M. Fotuhi).
Department of Clinical
Neurological Sciences,
University Hospital,
University of Western
Ontario, London, ON,
Canada (V. Hachinski).
Department of
Neurology, Case
Western Reserve
University, Cleveland,
OH, USA
(P. J. Whitehouse).
Correspondence:
M. Fotuhi, Center for
Memory and Brain
Health, The Sandra and
Malcolm Berman Brain
& Spine Institute, 5051
Greenspring Avenue,
Baltimore, MD 21209,
USA
mfotuhi@
lifebridgehealth.org
volume 5 | DECEMBER 2009 | 1
reviews
Key points
■■ Over the past 27 centuries, the perception of cognitive impairment with aging
has changed from a normal inevitable part of aging to being mostly attributable
to Alzheimer disease (AD)
■■ Alois Alzheimer was one of the first clinician–scientists to describe the
importance of vascular pathology and to de-emphasize the role of amyloid
plaques in brain atrophy and late-life dementia
■■ Clinicopathological studies have consistently shown that individuals over
80 years of age generally have ‘mixed’ pathologies (infarcts, plaques, tangles,
Lewy bodies and inflammation) rather than ‘pure AD’
■■ The size of the cortex and hippocampus—more than AD or any other single
pathological finding—correlates with the degrees of cognitive decline and
dementia in elderly individuals
formation of tangles, triggering widespread neuronal
dysfunction and dementia.9,10 Interest in this hypo­thesis
grew rapidly, and the term senile dementia was soon
changed to ‘senile dementia of Alzheimer’s type’ and,
eventually, simply to ‘Alzheimer disease’.
With increasing interest in plaques and tangles and
with the hope of finding a ‘cure’ for late-life dementia,
experts in neurology and psychiatry convened and established a set of criteria for a clinical diagnosis of AD.11,12 In
parallel, minimal microscopic criteria were established
for a postmortem histological diagnosis of AD. The
Khachaturian criteria, reflecting the opinion of a panel
of experts who met in 1984, attempted to standardize the
pathological diagnosis of AD on the basis of density of
senile plaques (both neuritic and diffuse) found in cortical areas.13 Higher densities of plaques were required
for a positive AD diagnosis as the age of the indivi­
duals increased from <50 years, through 50–75 years,
to >75 years.13 In 1991, the CERAD (Consortium to
Establish a Registry for AD) criteria were established to
provide further specificity for an AD diagnosis. Under
these criteria, densities of neuritic (not diffuse) plaques
above a defined normal value were assigned to three categories, A, B and C, with category C representing the
highest plaque density.14 CERAD used an age-related
plaque score, and the diagnostic categories also incorporated clinical information.
The CERAD criteria fulfilled an important need for
confirmation of an AD diagnosis in research and clinical centers around the world. However, this widely used
classification had two important limitations. First, the
pathological distinctions were based on the brains of 142
patients with dementia (average age 76 years) compared
with those of only eight much younger ‘control’ indivi­
duals (average age 65 years). If the authors had selected
patients and controls in their 80s, their cut-off criteria
could have been quite different. Second, the quantity
and distribution of neurofibrillary tangles—a prominent feature of AD—were not taken into account. These
limitations resulted in a great deal of disagreement, even
among neuropathologists viewing the same pathological
specimens.15
The Braak and Braak criteria (stages I–VI), which were
based on distribution and progression of neurofibrillary
tangles (not plaques) from limbic areas to frontal lobes,
showed a close correlation between stages of dementia
and the severity of pathological findings.16 The National
Institute on Aging–Reagan criteria, which were introduced in 1997, incorporated information about the severity of the burden of both plaques and tangles along with
clinical information regarding dementia, and removed
the criteria for age.17 AD pathology was highly likely to
be the underlying cause of dementia if both frequent
neuritic plaques (CERAD category C) and widespread
neurofibrillary tangles (Braak stage V and VI) were
present.
Clinical and pathological consensus guidelines have
been proposed for other forms of dementia, such as
s
w
e
i
v
D
e
E
R
T
e
r
C
u
E
t
R
a
N OR OF
C
O
N
R
U
P
■■ Appreciating the link between midlife risk factors and late-life size of the cortex
and hippocampus has serious implications for disease diagnosis, patient
management, and interpretation of research findings
■■ The dynamic polygon hypothesis provides a new framework for thinking about
aging and dementia that departs from the linear model proposed by the amyloid
cascade hypothesis
major academic institutions in Europe, included this case
report in his leading textbook of psychiatry and used
the term ‘Alzheimer’s disease’.1 Alzheimer himself did
not consider amyloid to be the primary cause of senile
dementia, and he wrote, “plaques are not the cause of
senile dementia, but only an accompanying feature of
senile involution of the central nervous system.”3
For much of the early 20th century, AD was con­sidered
to be a rare condition that affected young people with
presenile dementia. By contrast, ‘hardening of the blood
vessels’ was considered to be the main pathology for
cognitive decline in the last decades of life. In the 1940s
and 1950s, a group of psychiatrists, including David
Rothschild, promoted the idea that late-life dementia
was a consequence of society’s choice to isolate elderly
individuals and deprive them of meaningful inter­actions
with friends and relatives.4 David Wilson wrote, “lonesomeness, lack of responsibility, and a feeling of not being
wanted all increase the restricted view of life, which in
turn leads to restricted blood flow.”5 In 1974, the increasing realization that strokes can be the primary etiology for
brain atrophy and confusion in older patients led to the
formulation of the ‘multi-infarct dementia’ diagnosis.6
A gradual shift of focus from vascular issues to AD
pathology began with reported findings of extensive
amyloid plaque loads in the brains of elderly people
with dementia.7 The discovery of mutations in the genes
encoding γ‑secretase and amyloid precursor protein in
familial forms of early-onset AD put amyloid at the
center of the pathological processes of dementia, and
the amyloid cascade hypothesis attracted substantial
attention.8 This hypothesis proposes that aggregation
of amyloid‑β (Aβ) protein in the cortex (which begins
as toxic dimers and oligomers that later turn into diffuse
and then fibrillar ‘insoluble plaques’) triggers oxidative
injury and synaptic loss; these, in turn, bring about
hyperphosphorylation of tau protein, which leads to
2 | DECEMBER 2009 | volume 5
www.nature.com/nrneurol
reviews
dementia with Lewy bodies, vascular dementia, and
normal pressure hydrocephalus.18–21 The possibility that
cerebral amyloid angiopathy could contribute to dementia
through mechanisms other than paren­chymal AD pathology was also recognized.22 A major challenge acknow­
ledged by the consensus guidelines was the determination
of boundaries between normal aging and dementia among
elderly individuals (especially those beyond the age of
80 years), as those without considerable cognitive decline
often had some degree of pathology in their brains.15,23
Reflecting these challenges, one study showed that the
frequency of diagnosed cases of dementia in the same
patient population of 1,879 men and women over 65 years
of age varied by an order of magnitude (from 3.1% to
29.1%), depending on the clinical criteria used.24
s
w
e
i
v
D
e
E
R
T
e
r
C
u
E
t
R
a
N OR OF
C
O
N
R
U
P
1997–2007
The description of a clinical stage called mild cognitive
impairment (MCI) defined a major turning point in
dealing with the challenge of dichotomization of patients
into normal or dementia categories.25 Initially, the proposed diagnostic criteria for MCI required significant,
objectively measured memory loss that was corroborated
by the patient’s family. Further progress in refining the
definition of MCI came with the recognition that some
elderly individuals can have a nonamnestic presentation that leads to vascular or other forms of dementia.26
Pathology, imaging and cerebrospinal fluid studies all
pointed to MCI as a transitional stage along the trajectory of cognitive decline—a stage that could be targeted
for intervention in patients with a high likelihood of
developing dementia within 2–3 years.
Despite these new developments, the main focus
remained on AD. Throughout the 1990s, a consensus
had taken shape among clinicians and researchers in the
field that plaques and tangles eventually cause AD, and
that AD is the predominant cause of dementia among
the elderly.9 A major assumption that was made in 100s
of published studies, and which prevailed until recently,
was that most patients had either vascular dementia or
AD, but not both.
The Nun Study in 1997 reignited interest in the importance of adequate blood supply to the brain (Figure 1)
and the role of vascular disease and stroke in late-life
dementia.27,28 Examination of the brains of elderly nuns
revealed a distinct dissociation between the load of AD
plaques and tangles and the degree of cognitive impairment that was evident before their deaths. It became
clear from the Nun Study that lacunar strokes magnified the effects of any given load of AD pathology, and
vice versa. A large, multicenter, longitudinal study in
England and Wales published in 2001 also showed that
most patients with late-onset dementia had a mixture
of cerebro­vascular and AD‑type lesions.29 Patients who
had either mild subclinical (silent) AD pathology or mild
sub­clinical cerebrovascular disease seemed to remain free
of dementia for a longer period of time than those who
had a combination of these two pathologies.23,27,28,30
nature reviews | neurology Figure 1 | High density of blood vessels in the brain. To reveal the density of
cerebral blood vessels, the brain was injected with a plastic emulsion and the
parenchymal tissue was dissolved. As this specimen illustrates, the brain is a
highly vascular organ. Thus, vascular risk factors that impede adequate cerebral
flow can substantially impair all aspects of cognitive function with aging.
Permission obtained from Wolters Kluwer Health © Zlokovic, B. V. &
Apuzzo, M. L. J. Strategies to circumvent vascular barriers of the central nervous
system. Neurosurgery 43(4), 877–878 (1998).
Box 1 | Factors associated with cognitive function late in life
The summary below is semi-quantitative: performing a quantitative meta-analysis
for these associations remains challenging owing to marked heterogeneity in
selection of participants for longitudinal and interventional studies, and the wide
range of outcome measures selected in individual reports.28,30,32,33,49,52,53,58,60,65,67,
71,78,79,83–86,89,92,112,114–118
Factors associated with better cognitive function with aging
■■ Strong associations: education, walking (physical activity)
■■ Moderate association: leisure activities
■■ Mild associations: alcohol (one or two glasses per day), challenging occupation,
eating fish, eating fruit and vegetables
Factors associated with worse cognitive function with aging
■■ Strong associations: apolipoprotein E ε4 genotype, silent or large strokes,
midlife hypertension, obesity
■■ Moderate associations: depression, diabetes, excessive alcohol use, high
homocysteine levels, high midlife cholesterol levels, obstructive sleep apnea
■■ Mild associations: chronic stress, head trauma, impaired insulin response, low
folate and vitamin B12 levels, smoking
Numerous other clinical, pathological and radio­logical
findings have confirmed a close link between vascular
risk factors, the development of strokes (ranging from
microscopic to large), and late-onset cognitive decline
(Box 1).28,31,32 Some longitudinal epidemiological studies
that monitored participants from midlife to late life
revealed that a combination of risk factors could increase
the likelihood of dementia more than 16-fold.33 As vascular lesions could range from a few microscopic infarcts
or mild white matter changes to large strokes and marked
atrophy (with varying contribution to cognitive decline),
volume 5 | DECEMBER 2009 | 3
reviews
Box 2 | Neuropathological findings in individuals aged >80 years
Crystal et al. (2000)38
■■ Many patients >80 years with dementia do not meet pathological criteria for
Alzheimer disease (AD), dementia with Lewy bodies (DLB) or frontotemporal
dementia
■■ Incidence of non-AD pathology progressively increases beyond 70 years of age,
approaching 50% in nonagenarians
White et al. (2005)36
■■ Japanese American men—especially those diagnosed with AD—showed
considerable discrepancies between clinical diagnosis and pathological findings
absence of any vascular risk factors (as a way of excluding
individuals with coexisting vascular lesions), the number
of cases previously diagnosed with AD dropped by more
than 50%.34 In parallel, examination of brains of patients
with late-onset dementia revealed that the link between
plaques and tangles and symptoms of clinical dementia
was strong in patients younger than 75 years and poor
for those older than 90 years.35,36 Thus, skepticism grew
over the simplistic view that the common form of latelife dementia among the oldest old is primarily attributable to the accumulation of plaques and tangles in the
brain.35–40
s
w
e
i
v
D
e
E
R
T
e
r
C
u
E
t
R
a
N OR OF
C
O
N
R
U
P
■■ Late-life cognitive impairment and dementia often involve a combination of AD,
microvascular lesions, cortical Lewy bodies, hippocampal sclerosis, and diffuse
atrophy and/or neuronal loss
Schneider et al. (2007)41
■■ Patients with multiple pathologies were three times more likely to exhibit
dementia than those with only one pathology
■■ Mixed brain pathologies accounted for most dementia cases in patients aged
>80 years
Sonnen et al. (2007)45
■■ Independent correlates of dementia include Braak stage V or VI, more than two
infarcts, and Lewy bodies
■■ Interventions to reduce infarct risk might prevent or delay dementia onset
Haroutunian et al. (2008)46
■■ Individuals aged >80 years old show different neuropathological features of
dementia from septuagenarians
■■ Infarcts, DLB, hippocampal sclerosis, or factors yet to be identified, might
contribute to dementia in people aged >80 years
Savva et al. (2009)42
■■ Neocortical and hippocampal atrophy was a better predictor of dementia than
was AD pathology
■■ Therapeutic interventions targeting AD pathology might be effective for
septuagenarians but not octagenarians or nonagenarians
White (2009)43
■■ Certain lesion combinations, such as AD plus infarcts, were more closely
associated with dementia than were individual pathological lesions
■■ Infarcts were the dominant finding in many cases, perhaps because the
participants were elderly men
Schneider et al. (2009)44
■■ Odds ratios of clinically probable AD increase significantly when different
neuropathological lesions are combined
■■ Most elderly people with clinically diagnosed AD exhibit mixed pathologies
Erten-Lyons et al. (2009)47
■■ Large hippocampal and total brain volume allows elderly people to remain
cognitively healthy despite a high AD pathology burden
no consensus could be reached for a widely accepted
diagnosis of vascular dementia.20
With the realization that even small vascular lesions
have profound effects on the brain and can substantially
modify the link between AD pathology and dementia,
some researchers began to question the accuracy of AD
diagnoses in population studies. 34 In a retrospective
analysis, when ‘pure AD’ was defined as dementia in the
4 | DECEMBER 2009 | volume 5
2007 to the present day
An important study from 2007 confirmed that the load
of AD plaques and tangles in elderly individuals with
dementia could be similar to that found in cognitively
healthy individuals without dementia (30% and 24.2%,
respectively), and that the brains of patients with
dementia often had a combination of AD lesions, vascular pathology and Lewy bodies.41 A subsequent study
from the same group, along with several other reports,
showed that the presence of multiple pathologies significantly increases the likelihood of conversion from
cognitively normal to MCI, and from MCI to dementia (Box 2, Supplementary Table 1 online).36,38,42–47 The
odds ratio for a clinically probable AD diagnosis was 4.7
in the presence of AD pathology alone, but it increased
to 16.2 in the presence of a combination of AD patho­
logy, infarcts and Lewy bodies.44 Another study showed
that AD lesions fully account for dementia among the
‘young old’ (60–80 years) but not among the oldest old
(beyond 90 years).46 In the Honolulu–Asia Aging Study,
only 18.6% of elderly patients with a clinical diagnosis
of dementia had pure AD pathology.43 These and other
independent clinicopathological studies concluded that
late-life dementia reflects the convergence of several different pathological processes on the brain areas that are
important for memory and higher cognitive function;
that is, the cortex and hippocampus.48
Two studies reported in 2009 have highlighted the size
of the cortex and hippocampus as the main determinants of late-life dementia. Erton-Lyons and colleagues
analyzed the brains of 36 individuals (12 with normal
cognitive function and 24 with a diagnosis of AD before
death), all of whom met the standard pathological criteria for AD (Braak stage V or VI, and moderate to frequent neuritic plaques according to the CERAD criteria.
Larger cortical and hippocampal volumes were associated with preserved cognition, even in the presence of
a high burden of AD lesions.47 Another clinicopathological study showed that the load of neuritic plaques in
the hippocampus rises with each decade of life beyond
the age of 70 years in individuals without dementia,
but decreases in those with dementia.42 The degree of
atrophy in the cortex and hippocampus remained the
most consistent correlate of dementia in the last decades
of life.
www.nature.com/nrneurol
reviews
With the goal of finding effective strategies for prevention and treatment of cognitive impairment with aging,
new avenues of research are now focusing on all the
patho­logical and physiological processes that can potentially affect the cortex and hippocampus.47 Some midlife
risk factors are associated with marked late-life dementia
and with a smaller cortex and hippocampus.33 Extensive
research is now underway to elucidate the mechanisms
through which midlife factors might modulate the likelihood of dementia in the last decades of life (Box 1), and
to establish how vascular conditions might interact with
each other and with neurodegenerative processes such
as AD.49
Cortical atrophy
Plaques and tangles
Synucleopathy
White matter abnormalities?
Stroke
Hypoxia
Obstructive sleep apnea
Diabetes mellitus
Traumatic brain injury
Congestive heart failure
Hypertension
White matter abnormalities
Hypertension
Diabetes mellitus
Congestive heart failure
Kidney or liver disease?
Thyroid disease
Vitamin B12 deficiency
Lacunar stroke
Diabetes mellitus
Hypertension
High cholesterol?
Emboli
s
w
e
i
v
D
e
E
R
T
e
r
C
u
E
t
R
a
N OR OF
C
O
N
R
U
P
The dynamic polygon hypothesis
Late-life dementia, in contrast to early-onset AD, can
reflect damage to the brain by a wide range of vascular
and nonvascular factors (Figure 2). To varying degrees,
obesity, hypertension, diabetes, atrial fibrillation, high
cholesterol, congestive heart failure, inflammatory conditions (such as lupus), obstructive sleep apnea (OSA),
education, exercise, chronic stress, and depression can all
alter brain architecture transiently or permanently at the
cellular or macroscopic level (Figure 3).50–61 This broad
view, integrating brain function, cardiovascular function,
neuroplasticity, and eventual development of cognitive
impairment in late life, highlights a dynamic inter­
action between genetically determined, non­modifiable
pathological processes, and processes that are potentially
reversible (for example, environmental exposures). This
model, which we have termed the ‘dynamic polygon
hypothesis’, departs from a primary focus on plaques
and tangles (Figure 3). For example, small-vessel disease
and AD pathology are both linked to loss of neurons in
the CA1 area of the hippocampus.62 In turn, high blood
pressure, diabetes, obesity, smoking, and sedentary
lifestyle are prominent triggers for small-vessel disease
and might, therefore, influence the ultimate size of the
hippocampus. Moreover, an individual vascular factor
such as obesity could have nonlinear and heterogeneous
consequences that affect the brain through numerous
mediators, including hypertension, infarcts and white
matter changes, as well as increased sympathetic acti­
vity, interleukins, neurotrophins, growth factors, adipo­
cytokines, adipose-derived leptin, and satiety factors (see
below).63
Hypertension—a prominent feature of obesity—is
likely to lead to atrophy in the cortex and hippocampus
through vascular lesions; that is, atherosclerosis, white
matter changes, or infarcts.64 OSA, another condition
that is commonly associated with obesity, probably
causes marked cortical atrophy through chronic nocturnal cerebral hypoxia over a period of several decades.65
The severity of cortical atrophy in patients with OSA
approaches 18% in the frontal lobes, hippocampus, parahippocampal gyri, parietal cortex, and cingulate cortex—
many of the same cortical areas that are known to be
affected by AD pathology.66
nature reviews | neurology Hippocampal atrophy
Hippocampal sclerosis
Obstructive sleep apnea
White matter abnormalities?
Hypoxia
Plaques and tangles
Chronic depression or stress
Hypertension
Reduced cerebral blood flow
Diabetes mellitus
Hypertension
Cholesterol
Smoking
Cerebral amyloid angiopathy
Figure 2 | Factors that could cause brain atrophy and cognitive impairment. Blood
vessels are shown in red. Cortical and hippocampal volumes correlate well with
the degree of cognitive decline and dementia. Some processes lead directly to
atrophy in these structures, whereas others contribute to white matter
abnormalities and strokes (cortical or subcortical, small or large), and might
indirectly hasten volume loss in the cortex and hippocampus. Further
investigations will be required to ascertain the relative contributions of the various
processes—especially those indicated by question marks—to brain atrophy and
cognitive impairment.
Midlife obesity might lower the threshold for latelife dementia through mechanisms other than hypertension and/or hypoxic-induced brain atrophy due to
OSA.67 In a study that controlled for high blood pressure, myo­cardial infarction and strokes, these factors
did not fully explain the strong association between
midlife excessive abdominal fat accumulation and cognitive decline 30 years later.67 Other potential mediators
include insulin resistance, insulin-like growth factor,
inflammation, ghrelin, leptin, or other as yet unidentified hormones.68–70 High insulin levels could suppress the
insulin-degrading enzyme and lead to higher levels of Aβ
oligomers, as well as reducing Aβ clearance and increasing tau hyper­phosphorylation.69,70 In the setting of metabolic syndrome or diabetes, obesity can heighten levels
volume 5 | DECEMBER 2009 | 5
reviews
a The amyloid cascade hypothesis
b The dynamic polygon hypothesis
Missense mutations in APP, PS1 or PS2 genes
s
ue
oid
Increased Aβ42 production and accumulation
oli
po
pr
ot
E
path
y
ein
leino
Cognitive
impairment
Synu
c
Microglial and astrocytic activation
(for example, by complement factors, cytokines)
Ap
ion
tens
Subtle effects of Aβ oligomers on synapses
q
pla
er
Hyp
Aβ42 oligomerization and deposition as diffuse plaques
yl
Am
Neurofibrillary
tangles
s
w
e
i
v
D
e
E
R
T
e
r
C
u
E
t
R
a
N OR OF
C
O
N
R
U
P
Mild
Progressive synaptic and neuritic injury
Tangles
Widespread neuronal and neuritic dysfunction
and cell death with transmitter deficits
Alzheimer disease
Obe
sity
Severe
tes
Altered kinase and phosphatase activities
Intermediate
e
Diab
Altered neuronal ionic homeostasis; oxidative injury
Lo
w
c
re ogn
se iti
r ve ve
ion
at
m
am
Stroke
Infl
Figure 3 | Models to account for late-life cognitive impairment. a | According to the amyloid cascade hypothesis, a chain of
processes that begins with plaques, which in turn cause the formation of tangles, leads to synaptic loss and dementia. In
this model, no distinction is made between early-onset and late-life dementia. b | According to the dynamic polygon
hypothesis, early-onset dementia results from toxicity associated with aggregation of plaques and tangles (although not
necessarily in a linear fashion). Late-life dementia, on the other hand, is considered to be a more complex disease: a set
of pathological processes that affect the size of the cortex and hippocampus (for example, tauopathy, inflammation,
synucleinopathy, amyloid aggregation, and strokes) is interlinked with positive or negative consequences of environmental
exposures (for example, education, exercise, leisure activities, or obesity). In this model, plaques and tangles are two
components among a larger set of factors that modulate synaptic density and the size of the cortex and hippocampus,
and eventually determine the level of cognitive agility or frailty toward the end of life. More studies are needed to establish
which model best fits the existing data in this field. Abbreviations: Aβ, amyloid‑β; APP, amyloid precursor protein;
PS, presenilin.
of inflammatory processes in the brain and, either independently or in conjunction with degenerative processes,
modulate the size of the cortex and hippocampus.58,71 The
obesity–dementia link, therefore, illustrates how various
vascular and nonvascular processes might interact in a
dynamic and complex network of factors encompassing
both genes and environment.
The apolipoprotein E ε4 (APOE ε4) genotype, a major
genetic susceptibility factor for AD, increases the risk of
dementia between twofold and 12-fold.72 The presence of
APOE ε4 lowers the efficiency of cellular repair mechanisms, reduces the clearance of plaques and tangles in the
brain, and decreases gray matter volume.73,74 The impact
of APOE ε4 on the brain is not limited to plaques and
tangles, however, as the Apo‑E4 protein can interact with
and directly modify the severity of vascular conditions
such as hypercholesteremia. Apo‑E4 has also been shown
to directly interact with the LDL cholesterol receptor.75
Other forms of Apo‑E might also have roles in both
AD and non-AD processes. The APOE ε2 genotype
is believed to exert neuroprotective effects through a
lowering of levels of Aβ in the brain. A recent study,
however, showed that elderly individuals beyond the age
of 90 years who possess APOE ε2 tend to remain cognitively intact even in the presence of a high burden of
plaques and tangles, suggesting involvement of non-AD-
6 | DECEMBER 2009 | volume 5
related neuroprotective mechanisms against dementia.76
Thus, presence of the APOE ε4 or APOE ε2 allele might
alter the course of cognitive decline with aging through
changes in levels of both AD and vascular injury, as well
as through modifications in compensatory repair mechanisms that deal with both types of pathology.
In contrast to early-onset AD, which cannot be modified with any known interventions, late-life dementia
might be preventable. Some preliminary—and still
controversial—findings indicate that a number of protective factors, including exercise, education, participating in brain-stimulating activity, having a cognitively
challenging occupation, eating an antioxidant-rich diet,
and consuming fish or omega‑3 fatty acid supplements,
are associated with improved cognitive function and a
reduced risk of late-life dementia (Box 1).30 These factors
are believed to increase synaptic density in the brain (that
is, create stronger cognitive reserve), perhaps in part
through angiogenesis and in part through increasing
levels of brain-derived neurotrophic factor (BDNF).77,78
In animals, exercise selectively increases BDNF gene
expression in the hippocampus and reduces the load of
amyloid plaques throughout the brain.79,80
Results from studies of neuroplasticity in the adult
animal brain are beginning to be replicated in humans.
Sensitive MRI measurements reveal that the size of the
www.nature.com/nrneurol
reviews
human cortex and hippocampus can expand significantly with exercise or intense brain stimulation. In a
placebo-controlled study, healthy elderly people who
participated in a walking program 3 days a week for
6 months experienced a 3% increase in cortical brain
volume in their frontal lobes, as determined by MRI
findings before and after the exercise program. 81,82 In
medical students preparing for their national certification examinations, intensive brain stimulation over a
period of 3 months was shown to increase the volume of
the cortex and hippocampus.83 These observations might
partly account for resistance to injury triggered by AD
pathology, which is observed in individuals with high
levels of fitness and cognitive reserve; these individuals
could have optimal cerebral blood flow in their brains
and a relatively high density of synapses in their cortex
and hippocampus.84,85
In summary, the primary focus on AD pathology to
account for late-life dementia is being superseded by
a focus on understanding potentially modifiable processes.86 According to the dynamic polygon hypothesis,
a balance of positive and negative environmental factors,
together with a balance of positive and negative genetic
factors, seems to affect the brain throughout early life
and midlife to determine the degree of cognitive agility
or impairment in late life (Box 1, Figure 2).84,85 These
factors increase or decrease cerebral blood flow, oxidative
stress, inflammation, insulin-signaling components, size
and frequency of infarcts, and concentrations of growth
factors, cortisol or other hormones.
Preliminary reports suggest that the load of amyloid
plaques, which is determined to some extent by genetic
background, can potentially be altered by environ­
mental factors such as exercise, traumatic brain injury or
diet.80,87–89 In animal studies, consumption of apple juice
or curcumin seemed to lower amyloid levels.90,91 Thus,
like the degree of microvascular disease in the brain,
amyloid levels might depend on lifestyle choices. These
observations have provided a strong impetus to establish
the profile of risk factors for dementia in late life and
to initiate early preventive strategies in individuals with
a high likelihood of developing cognitive decline with
aging.92 These preventive strategies would aim to modify
both vascular and AD pathology in the brain through
changes in lifestyle and use of disease-modifying drugs.
PIB imaging data with cerebrospinal fluid findings. 94
Standard MRI techniques have enabled us to establish
that hippocampal volume is an excellent predictor of
further deterioration in patients with MCI and dementia.95,96 New MRI techniques, such as diffusion tensor
imaging, are beginning to reveal the degree and relevance
of white matter changes with aging.97
More than 100 clinical trials of approaches to prevent
and treat patients with varying degrees of cognitive
impairment are currently underway.98,99 Drugs being
tested include immune-related medications (for example,
immunoglobulin or vaccines), inhibitors of amyloid and
tau, and nerve-growth-factor-like agents.99,100 Despite the
fact that an initial active immunization trial to reduce
levels of amyloid in patients with dementia was stopped
owing to encephalitic complications, and the preliminary
(and incomplete) results were disappointing, passive
immunization clinical trials are still in progress. 101,102
Research is also underway to detect ‘cognitively normal’
individuals at risk of late-life dementia at the pre­
symptomatic stage, and to determine the ideal diseasemodifying medications for these individuals.60 Treatment
of vascular risk factors is associated with a reduced rate
of cognitive decline, and preventive strategies in this area
are starting to move from ideas and suggestions to reallife recommendations for clinical practice.92 The ultimate
goal is to determine early-life or midlife interventions,
such as factors that enhance cognitive reserve and synap­
tic density, that would enable people to remain cognitively intact in their 80s and 90s, even if they develop a
high load of AD pathology in their brains (Figure 3b).
s
w
e
i
v
D
e
E
R
T
e
r
C
u
E
t
R
a
N OR OF
C
O
N
R
U
P
Future prospects
Ongoing trials
Over the past two decades, important refinements in
defining the pathophysiology of dementia have paved
the way for developing effective preventive and treatment strategies. In particular, our understanding of the
factors that could cause brain atrophy in late life has
expanded substantially over the past 2–3 years. New
imaging techniques, such as PET scans using 11C-labeled
Pittsburgh compound B (PIB) have unveiled the distribution of amyloid in the brain in patients with or without
dementia.93 Studies are now in progress that correlate
nature reviews | neurology Remaining questions
Serial PIB and MRI studies in normal individuals and
those with MCI or AD demonstrate a clear dissociation
between the annual rate of amyloid deposition and the
rate of brain atrophy and neurodegeneration, consistent
with previous observations that progression of clinical
symptoms in dementia is not coupled to amyloid deposition.103,104 Consequently, is amyloid still a valid target
for the treatment of elderly individuals with late-life
dementia and, if so, should research focus on the natural
compensatory mechanisms that confront amyloid, on
amyloid itself, or on both?105 Alternatively, should the
focus shift toward the dissolution of tau aggregates, given
that the density of neurofibrillary tangles correlates more
closely with the degree of cognitive impairment than
does amyloid pathology?39
Strong evidence in support of the amyloid cascade
hypothesis links toxic soluble amyloid dimers and oligomers to AD.106 However, attempts to demonstrate a
cascade process from amyloid aggregation to tanglerelated neuronal dysfunction have been disappointing, and no convincing causal link has been established
between plaques and tangles. 102 These findings have
called the amyloid cascade into question, and investigators must now consider how, and indeed whether, this
hypothesis should be tested further.102,107–111
volume 5 | DECEMBER 2009 | 7
reviews
A body of literature—albeit controversial—suggests
that anti-inflammatory and antioxidant medications can
lead to better cognitive function and a lower risk of cognitive impairment with aging.112 Future studies should
address whether inflammation is a common denominator in AD, dementia with Lewy bodies, white matter
changes and infarcts, and whether late-life dementia is a
primary neuroinflammatory condition that is aggravated
by other coexisting pathologies.
Anot her imp or t ant issue to address is t he
­relationship—if any—between late-life dementia and
early-onset AD. Is the common form of late-life dementia simply an extension of early-onset AD, or is it a separate condition, perhaps triggered by genes and proteins
that have yet to be discovered?40,46 Atrophy in the cortex
and hippocampus correlates better with the severity and
progression of late-life dementia than do white matter
changes, infarcts, plaques, tangles or Lewy bodies.47 The
pathogenetic basis of this atrophy is currently unclear:
could it result from processes other than strokes, AD,
inflammation, and Lewy body pathology? Given the
large number of clinicopathological studies that point to
the presence of multiple classes of pathology in brains of
the oldest old (with or without dementia), a case could
be made for re-evaluating the diagnostic criteria for AD
in patients beyond the age of 80 years.
Numerous midlife risk factors have been associated
with late-life dementia, ranging from early-life education, smoking, choice of hobbies and head trauma, to the
presence of medical conditions such as obesity (Box 1).
The pathophysiological mechanisms that underlie these
associations, and the factors that are most relevant for
identifying targets for early intervention, remain to
be determined. Future research should also focus on
which biomarkers are the best candidates for detecting presymptomatic patients who are at risk of late-life
dementia.
Given that a number of environmental risk factors
have been implicated in late-life dementia, and consider­
ing that rates of obesity and hypertension are rising at
a rapid rate among children, efforts to prevent dementia should perhaps start early in life. Numerous studies
have examined a possible role for omega‑3 fatty acids in
reducing the risk of dementia, but the results obtained
to date have been heterogeneous.89 One explanation for
the marked variation in findings from dozens of studies
in this field—and perhaps the explanation for the failure
of most clinical trials in patients with AD—could be the
selection and monitoring of participants with various
brain pathologies, all of whom were diagnosed with AD.
Given the observed heterogeneity of the pathological
process in patients with cognitive decline, candidates for
clinical trials should perhaps be selected more rigorously,
and be subdivided into groups with primary AD, primary
vascular pathology, or primary mixed pathology.
The last question is one of terminology. Some
researchers consider the word ‘dementia’ to be obsolete
and derogatory.113 Should we replace this diagnostic
terminology with a more respectful label such as ‘cognitive impairment’? The progressive deterioration in
cognitive function might be labeled on a scale ranging
from MCI, which already has its own established criteria, through intermediate cognitive decline (patients
who have developed difficulty in performing instrumental activities of daily living such as shopping), to severe
cognitive impairment (patients who have developed difficulties performing their basic activities of daily living
such as managing personal hygiene).
s
w
e
i
v
D
e
E
R
T
e
r
C
u
E
t
R
a
N OR OF
C
O
N
R
U
P
8 | DECEMBER 2009 | volume 5
Conclusions
The dominant conceptual views of late-life memory
loss and confusion have shifted considerably throughout history. These symptoms were considered ‘normal’
as early as 700 BC, as signs of being ‘possessed by evil’
in the early Renaissance period, as evidence of ‘harden­
ing of blood vessels’ throughout most of the 20th
century, and as manifestations of AD since the 1990s.
Clinicopathological studies conducted over the past few
years agree that most individuals with cognitive impairment over the age of 80 years have a mixture of several
coexisting abnormalities, and only a small proportion
have pure pathology (for example, dementia with Lewy
bodies, AD, or hippocampal sclerosis) in their brains.
Technological advances in brain imaging, along with
advances in the field of neuroscience, have opened up
new possibilities for studying the brain with aging, and
have provided an opportunity for researchers to ask
more-definitive questions. An enormous amount of
progress has been made, but more research is required
before specific recommendations to prevent late-life
dementia can be formulated.
Alois Alzheimer was one of the first scientists to extensively describe the importance of vascular lesions in
brain atrophy in late-life dementia (and to de-emphasize
the relevance of amyloid plaques). It is noteworthy that
a century later, reduction of vascular risk factors (along
with improvement of physical and cognitive fitness)
remains the most reasonable recommendation that we
can offer to members of the public who strive toward
better brain health and successful aging.28,33,92,114,115
Review criteria
MEDLINE was searched for articles published in
English from January 1980 to August 2009, with the
following keywords: “dementia”, “cognitive impairment”,
“memory”, “Alzheimer disease”, “amyloid hypothesis”,
“aging” and “clinicopathologic”. Abstracts were reviewed,
and papers with a focus on the link between clinical
manifestation of cognitive decline and diagnostic
criteria for dementia, as well as those with an emphasis
on historical development of concepts in the field of
dementia, were further analyzed in detail. In addition,
the reference sections of these articles, along with
relevant chapters in standard neuropathology textbooks,
were consulted.
www.nature.com/nrneurol
reviews
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
Berchtold, N. C. & Cotman, C. W. Evolution in the
conceptualization of dementia and Alzheimer’s
disease: Greco-Roman period to the 1960s.
Neurobiol. Aging 19, 173–189 (1998).
Mast, H., Tatemichi, T. K. & Mohr, J. P. Chronic
brain ischemia: the contributions of Otto
Binswanger and Alois Alzheimer to the
mechanisms of vascular dementia. J. Neurol. Sci.
132, 4–10 (1995).
Alzheimer, A. On peculiar cases of disease at
higher age [German]. Neurologie und Psychiatrie
4, 256–286 (1911).
Ballenger, J. F. Progress in the history of
Alzheimer’s disease: the importance of context.
J. Alzheimers Dis. 9, 5–13 (2006).
Wilson, D. C. The pathology of senility. Am. J.
Psychiatry 111, 902–906 (1955).
Hachinski, V. C., Lassen, N. A. & Marshall, J.
Multi-infarct dementia. A cause of mental
deterioration in the elderly. Lancet 2, 207–210
(1974).
Blessed, G., Tomlinson, B. E. & Roth, M. The
association between quantitative measures of
dementia and of senile change in the cerebral
grey matter of elderly subjects. Br. J. Psychiatry
114, 797–811 (1968).
Hardy, J. A. & Higgins, G. A. Alzheimer’s disease:
the amyloid cascade hypothesis. Science 256,
184–185 (1992).
Cummings, J. L. Alzheimer’s disease. N. Engl. J.
Med. 351, 56–67 (2004).
Hardy, J. & Selkoe, D. J. The amyloid hypothesis of
Alzheimer’s disease: progress and problems on
the road to therapeutics. Science 297, 353–356
(2002).
McKhann, G. et al. Clinical diagnosis of
Alzheimer’s disease: report of the NINCDS-ADRDA
Work Group under the auspices of Department of
Health and Human Services Task Force on
Alzheimer’s Disease. Neurology 34, 939–944
(1984).
American Psychiatric Association. Diagnostic and
Statistical Manual of Mental Disorders, 4th edn
(American Psychiatric Association, Washington,
D. C., 1994).
Khachaturian, Z. S. Diagnosis of Alzheimer’s
disease. Arch. Neurol. 42, 1097–1105 (1985).
Mirra, S. S. et al. The Consortium to Establish a
Registry for Alzheimer’s Disease (CERAD). Part II.
Standardization of the neuropathologic
assessment of Alzheimer’s disease. Neurology
41, 479–486 (1991).
Jellinger, K. A. Criteria for the neuropathological
diagnosis of dementing disorders: routes out of
the swamp? Acta Neuropathol. 117, 101–110
(2009).
Braak, H. & Braak, E. Neuropathological stageing
of Alzheimer-related changes. Acta Neuropathol.
82, 239–259 (1991).
[No authors listed] Consensus recommendations
for the postmortem diagnosis of Alzheimer’s
disease. The National Institute on Aging, and
Reagan Institute Working Group on Diagnostic
Criteria for the Neuropathological Assessment of
Alzheimer’s Disease. Neurobiol. Aging 18 (4
Suppl.), S1–S2 (1997).
McKeith, I. G. Dementia with Lewy bodies. Br. J.
Psychiatry 180, 144–147 (2002).
McKeith, I. G. et al. Consensus guidelines for the
clinical and pathologic diagnosis of dementia with
Lewy bodies (DLB): report of the Consortium on
DLB International Workshop. Neurology 47,
1113–1124 (1996).
Hachinski, V. et al. National Institute of
Neurological Disorders and Stroke-Canadian
Stroke Network vascular cognitive impairment
harmonization standards. Stroke 37, 2220–2241
(2006).
21. Relkin, N., Marmarou, A., Klinge, P., Bergsneider,
M. & Black, P. M. Diagnosing idiopathic normalpressure hydrocephalus. Neurosurgery 57 (3
Suppl.), S4–S16 (2005).
22. Yoshimura, M. et al. Dementia in cerebral amyloid
angiopathy: a clinicopathological study. J. Neurol.
239, 441–450 (1992).
23. Nagy, Z. et al. The effects of additional pathology
on the cognitive deficit in Alzheimer disease. J.
Neuropathol. Exp. Neurol. 56, 165–170 (1997).
24. Erkinjuntti, T., Ostbye, T., Steenhuis, R. &
Hachinski, V. The effect of different diagnostic
criteria on the prevalence of dementia. N. Engl. J.
Med. 337, 1667–1674 (1997).
25. Petersen, R. C. et al. Mild cognitive impairment:
clinical characterization and outcome. Arch.
Neurol. 56, 303–308 (1999).
26. Petersen, R. C. & Negash, S. Mild cognitive
impairment: an overview. CNS Spectr. 13, 45–53
(2008).
27. Snowdon, D. A. et al. Brain infarction and the
clinical expression of Alzheimer disease. The Nun
Study. JAMA 277, 813–817 (1997).
28.Viswanathan, A., Rocca, W. A. & Tzourio, C.
Vascular risk factors and dementia: how to move
forward? Neurology 72, 368–374 (2009).
29. Neuropathology Group. Medical Research
Council Cognitive Function and Aging Study.
Pathological correlates of late-onset dementia
in a multicentre, community-based population in
England and Wales. Neuropathology Group of the
Medical Research Council Cognitive Function and
Ageing Study (MRC CFAS). Lancet 357, 169–175
(2001).
30. Fotuhi, M. Preserving memory: tips to help baby
boomers stay in the game. Practical Neurology
March/April, [ED: is this volume number
correct?] 34–40 (2009).
31. Troncoso, J. C. et al. Effect of infarcts on dementia
in the Baltimore longitudinal study of aging. Ann.
Neurol. 64, 168–176 (2008).
32. Kivipelto, M. et al. Obesity and vascular risk
factors at midlife and the risk of dementia and
Alzheimer disease. Arch. Neurol. 62, 1556–1560
(2005).
33. Kivipelto, M. et al. Risk score for the prediction of
dementia risk in 20 years among middle aged
people: a longitudinal, population-based study.
Lancet Neurol. 5, 735–741 (2006).
34. Aguero-Torres, H., Kivipelto, M. & von Strauss, E.
Rethinking the dementia diagnoses in a
population-based study: what is Alzheimer’s
disease and what is vascular dementia? A study
from the Kungsholmen project. Dement. Geriatr.
Cogn. Disord. 22, 244–249 (2006).
35. Prohovnik, I. et al. Dissociation of
neuropathology from severity of dementia in lateonset Alzheimer disease. Neurology 66, 49–55
(2006).
36. White, L. et al. Recent clinical–pathologic
research on the causes of dementia in late life:
update from the Honolulu–Asia Aging Study.
J. Geriatr. Psychiatry Neurol. 18, 224–227
(2005).
37. Schmitt, F. A. et al. “Preclinical” AD revisited:
neuropathology of cognitively normal older adults.
Neurology 55, 370–376 (2000).
38. Crystal, H. A. et al. The relative frequency of
“dementia of unknown etiology” increases with
age and is nearly 50% in nonagenarians. Arch.
Neurol. 57, 713–719 (2000).
39. Nelson, P. T., Braak, H. & Markesbery, W. R.
Neuropathology and cognitive impairment in
Alzheimer disease: a complex but coherent
40.
41.
42.
43.
relationship. J. Neuropathol. Exp. Neurol. 68, 1–14
(2009).
Korczyn, A. D. The amyloid cascade hypothesis.
Alzheimers Dement. 4, 176–178 (2008).
Schneider, J. A., Arvanitakis, Z., Bang, W. &
Bennett, D. A. Mixed brain pathologies account
for most dementia cases in community-dwelling
older persons. Neurology 69, 2197–2204
(2007).
Savva, G. M. et al. Age, neuropathology, and
dementia. N. Engl. J. Med. 360, 2302–2309
(2009).
White, L. Brain lesions at autopsy in older
Japanese-American men as related to cognitive
impairment and dementia in final years of life:
a summary report from the Honolulu–Asia Aging
Study. J. Alzheimers Dis. doi:10.3233/
JAD‑2009‑1178.
Schneider, J. A., Arvanitakis, Z., Leurgans, S. E. &
Bennett, D. A. The neuropathology of probable
Alzheimer disease and mild cognitive impairment.
Ann. Neurol. 66, 200–208 (2009).
Sonnen, J. A. et al. Pathological correlates of
dementia in a longitudinal, population-based
sample of aging. Ann. Neurol. 62, 406–13
(2007).
Haroutunian, V. et al. Role of the neuropathology
of Alzheimer disease in dementia in the oldestold. Arch. Neurol. 65, 1211–1217 (2008).
Erten-Lyons, D. et al. Factors associated with
resistance to dementia despite high Alzheimer
disease pathology. Neurology 72, 354–360
(2009).
Jagust, W. J. et al. Neuropathological basis of
magnetic resonance images in aging and
dementia. Ann. Neurol. 63, 72–80 (2008).
Helzner, E. P. et al. Contribution of vascular risk
factors to the progression in Alzheimer disease.
Arch. Neurol. 66, 343–348 (2009).
Korf, E. S., White, L. R., Scheltens, P. &
Launer, L. J. Midlife blood pressure and the risk of
hippocampal atrophy: the Honolulu Asia Aging
Study. Hypertension 44, 29–34 (2004).
Du, A. T. et al. Age effects on atrophy rates of
entorhinal cortex and hippocampus. Neurobiol.
Aging 27, 733–740 (2006).
Sapolsky, R. M. Chickens, eggs and hippocampal
atrophy. Nat. Neurosci. 5, 1111–1113 (2002).
Sapolsky, R. M. Depression, antidepressants, and
the shrinking hippocampus. Proc. Natl Acad. Sci.
USA 98, 12320–12322 (2001).
Knecht, S. et al. Atrial fibrillation in stroke-free
patients is associated with memory impairment
and hippocampal atrophy. Eur. Heart J. 29,
2125–2132 (2008).
den Heijer, T. et al. Type 2 diabetes and atrophy of
medial temporal lobe structures on brain MRI.
Diabetologia 46, 1604–1610 (2003).
Appenzeller, S., Carnevalle, A. D., Li, L. M.,
Costallat, L. T. & Cendes, F. Hippocampal atrophy
in systemic lupus erythematosus. Ann. Rheum.
Dis. 65, 1585–1589 (2006).
Tate, D. F. & Bigler, E. D. Fornix and hippocampal
atrophy in traumatic brain injury. Learn. Mem. 7,
442–446 (2000).
Biessels, G. J., De Leeuw, F. E., Lindeboom, J.,
Barkhof, F. & Scheltens, P. Increased cortical
atrophy in patients with Alzheimer’s disease and
type 2 diabetes mellitus. J. Neurol. Neurosurg.
Psychiatry 77, 304–307 (2006).
Sheline, Y. I., Gado, M. H. & Kraemer, H. C.
Untreated depression and hippocampal volume
loss. Am. J. Psychiatry 160, 1516–1518 (2003).
Barnes, D. E. et al. Predicting risk of dementia in
older adults. The late-life dementia risk index.
Neurology 73, 173–179 (2009).
s
w
e
i
v
D
e
E
R
T
e
r
C
u
E
t
R
a
N OR OF
C
O
N
R
U
P
nature reviews | neurology 44.
45.
46.
47.
48.
49.
50.
51.
52.
53.
54.
55.
56.
57.
58.
59.
60.
volume 5 | DECEMBER 2009 | 9
reviews
61. Jefferson, A. L. et al. Lower cardiac output is
associated with greater white matter
hyperintensities in older adults with
cardiovascular disease. J. Am. Geriatr. Soc. 55,
1044–1048 (2007).
62. Kril, J. J., Patel, S., Harding, A. J. & Halliday, G. M.
Patients with vascular dementia due to
microvascular pathology have significant
hippocampal neuronal loss. J. Neurol. Neurosurg.
Psychiatry 72, 747–751 (2002).
63. Gustafson, D. Adiposity indices and dementia.
Lancet Neurol. 5, 713–720 (2006).
64. Wiseman, R. M. et al. Hippocampal atrophy, whole
brain volume, and white matter lesions in older
hypertensive subjects. Neurology 63, 1892–1897
(2004).
65. Minoguchi, K. et al. Silent brain infarction and
platelet activation in obstructive sleep apnea. Am.
J. Respir. Crit. Care Med. 175, 612–617 (2007).
66. Macey, P. M. et al. Brain morphology associated
with obstructive sleep apnea. Am. J. Respir. Crit.
Care Med. 166, 1382–1387 (2002).
67. Whitmer, R. A. et al. Central obesity and increased
risk of dementia more than three decades later.
Neurology 71, 1057–1064 (2008).
68. Diano, S. et al. Ghrelin controls hippocampal
spine synapse density and memory performance.
Nat. Neurosci. 9, 381–388 (2006).
69. LeRoith, D. Insulin-like growth factors and the
brain. Endocrinology 149, 5951 (2008).
70. Neumann, K. F. et al. Insulin resistance and
Alzheimer’s disease: molecular links & clinical
implications. Curr. Alzheimer Res. 5, 438–447
(2008).
71. Yaffe, K. et al. The metabolic syndrome,
inflammation, and risk of cognitive decline. JAMA
292, 2237–2242 (2004).
72. Farrer, L. A. et al. Effects of age, sex, and ethnicity
on the association between apolipoprotein E
genotype and Alzheimer disease. A meta-analysis.
APOE and Alzheimer Disease Meta Analysis
Consortium. JAMA 278, 1349–1356 (1997).
73. Tiraboschi, P. et al. Impact of APOE genotype on
neuropathologic and neurochemical markers of
Alzheimer disease. Neurology 62, 1977–1983
(2004).
74. Drzezga, A. et al. Effect of APOE genotype on
amyloid plaque load and gray matter volume in
Alzheimer disease. Neurology 72, 1487–1494
(2009).
75. Cheng, D. et al. Functional interaction between
APOE4 and LDL receptor isoforms in Alzheimer’s
disease. J. Med. Genet. 42, 129–131 (2005).
76. Berlau, D. J., Corrada, M. M., Head, E. &
Kawas, C. H. APOE ε2 is associated with intact
cognition but increased Alzheimer pathology in
the oldest old. Neurology 72, 829–834 (2009).
77. Cotman, C. W., Berchtold, N. C. & Christie, L. A.
Exercise builds brain health: key roles of growth
factor cascades and inflammation. Trends
Neurosci. 30, 464–472 (2007).
78. Helzner, E. P., Scarmeas, N., Cosentino, S.,
Portet, F. & Stern, Y. Leisure activity and cognitive
decline in incident Alzheimer disease. Arch.
Neurol. 64, 1749–1754 (2007).
79. Neeper, S. A., Gomez-Pinilla, F., Choi, J. &
Cotman, C. W. Physical activity increases mRNA
for brain-derived neurotrophic factor and nerve
growth factor in rat brain. Brain Res. 726, 49–56
(1996).
80. Adlard, P. A., Perreau, V. M., Pop, V. &
Cotman, C. W. Voluntary exercise decreases
amyloid load in a transgenic model of Alzheimer’s
disease. J. Neurosci. 25, 4217–4221 (2005).
81. Erickson, K. I. et al. Training-induced plasticity in
older adults: effects of training on hemispheric
asymmetry. Neurobiol. Aging 28, 272–283 (2007).
82. Colcombe, S. J. et al. Aerobic exercise training
increases brain volume in aging humans.
J. Gerontol. A Biol. Sci. Med. Sci. 61, 1166–1170
(2006).
83. Draganski, B. et al. Temporal and spatial
dynamics of brain structure changes during
extensive learning. J. Neurosci. 26, 6314–6317
(2006).
84. Drachman, D. A. Aging of the brain, entropy, and
Alzheimer disease. Neurology 67, 1340–1352
(2006).
85. Drachman, D. A. Nature or nurture: education and
the trajectory of declining brain function with age
and Alzheimer disease. Neurology 70, 1725–
1727 (2008).
86. McGurn, B., Deary, I. J. & Starr, J. M. Childhood
cognitive ability and risk of late-onset Alzheimer
and vascular dementia. Neurology 71,
1051–1056 (2008).
87. Ambrée, O. et al. Reduction of amyloid angiopathy
and Aβ plaque burden after enriched housing in
TgCRND8 mice: involvement of multiple
pathways. Am. J. Pathol. 169, 544–552 (2006).
88. Hartman, R. E. et al. Pomegranate juice
decreases amyloid load and improves behavior in
a mouse model of Alzheimer’s disease. Neurobiol.
Dis. 24, 506–515 (2006).
89. Fotuhi, M., Mohassel, P. & Yaffe, K. Fish
consumption, long-chain omega‑3 fatty acids and
risk of cognitive decline or Alzheimer disease:
a complex association. Nat. Clin. Pract. Neurol. 5,
140–152 (2009).
90. Yang, L. et al. Inhibition of the expression of
prostate specific antigen by curcumin [Chinese].
Yao Xue Xue Bao 40, 800–803 (2005).
91. Chan, A. & Shea, T. B. Dietary supplementation
with apple juice decreases endogenous amyloid‑β
levels in murine brain. J. Alzheimers Dis. 16,
167–171 (2009).
92. Kivipelto, M. & Solomon, A. Preventive neurology:
on the way from knowledge to action. Neurology
73, 168–169 (2009).
93. Aizenstein, H. J. et al. Frequent amyloid
deposition without significant cognitive
impairment among the elderly. Arch. Neurol. 65,
1509–1517 (2008).
94. Mathis, C. A. Amyloid imaging fundings from
multicenter studies. Alzheimer’s & Dementia 5,
IC‑S1‑01 (2009).
95. van der Flier, W. M. & Scheltens, P. Alzheimer
disease: hippocampal volume loss and
Alzheimer disease progression. Nat. Rev. Neurol.
5, 361–362 (2009).
96. Driscoll, I. et al. Longitudinal pattern of regional
brain volume change differentiates normal aging
from MCI. Neurology 72, 1906–1913 (2009).
97. Salmond, C. H. et al. Diffusion tensor imaging in
chronic head injury survivors: correlations with
learning and memory indices. Neuroimage 29,
117–124 (2006).
98. Cummings, J. L., Doody, R. & Clark, C. Diseasemodifying therapies for Alzheimer disease:
challenges to early intervention. Neurology 69,
1622–1634 (2007).
99. Salloway, S. & Correia, S. Alzheimer disease: time
to improve its diagnosis and treatment. Cleve.
Clin. J. Med. 76, 49–58 (2009).
100.Fillit, H., Hess, G., Hill, J., Bonnet, P. & Toso, C.
IV immunoglobulin is associated with a reduced
risk of Alzheimer disease and related disorders.
Neurology 73, 180–185 (2009).
101.Holmes, C. et al. Long-term effects of Aβ42
immunisation in Alzheimer’s disease: follow-up of
a randomised, placebo-controlled phase I trial.
Lancet 372, 216–223 (2008).
102.Hardy, J. The amyloid hypothesis for Alzheimer’s
disease: a critical reappraisal. J. Neurochem.
110, 1129–1134 (2009).
103.Jack, C. R., Jr et al. Serial PIB and MRI in normal,
mild cognitive impairment and Alzheimer’s
disease: implications for sequence of
pathological events in Alzheimer’s disease. Brain
132, 1355–1365 (2009).
104.Giannakopoulos, P. et al. Tangle and neuron
numbers, but not amyloid load, predict cognitive
status in Alzheimer’s disease. Neurology 60,
1495–1500 (2003).
105.Jagust, W. Will neuroimaging help us understand
Alzheimer’s disease? Alzheimer’s & Dementia 5,
IC‑PL‑01 (2009).
106.Shankar, G. M. et al. Amyloid‑β protein dimers
isolated directly from Alzheimer’s brains impair
synaptic plasticity and memory. Nat. Med. 14,
837–842 (2008).
107.Lee, H. G. et al. Amyloid‑β in Alzheimer disease:
the null versus the alternate hypotheses.
J. Pharmacol. Exp. Ther. 321, 823–829 (2007).
108.Seabrook, G. R., Ray, W. J., Shearman, M. &
Hutton, M. Beyond amyloid: the next generation
of Alzheimer’s disease therapeutics. Mol. Interv.
7, 261–270 (2007).
109.Abbott, A. Neuroscience: the plaque plan. Nature
456, 161–164 (2008).
110.Small, S. A. & Duff, K. Linking Aβ and tau in lateonset Alzheimer’s disease: a dual pathway
hypothesis. Neuron 60, 534–542 (2008).
111.Williams, M. Progress in Alzheimer’s disease drug
discovery: an update. Curr. Opin. Investig. Drugs
10, 23–34 (2009).
112.Fotuhi, M. et al. Better cognitive performance in
elderly taking antioxidant vitamins E and C
supplements in combination with nonsteroidal
anti-inflammatory drugs: the Cache County Study.
Alzheimers Dement. 4, 223–227 (2008).
113.Trachtenberg, D. I. & Trojanowski, J. Q. Dementia:
a word to be forgotten. Arch. Neurol. 65, 593–595
(2008).
114.Hachinski, V. World Stroke Day 2008: “little
strokes, big trouble”. Stroke 39, 2407–2420
(2008).
115.Hachinski, V. Shifts in thinking about dementia.
JAMA 300, 2172–2173 (2008).
116.Knopman, D. S. Go to the head of the class to
avoid vascular dementia and skip diabetes and
obesity. Neurology 71, 1046–1047 (2008).
117.Kivipelto, M. & Solomon, A. Cholesterol as a risk
factor for Alzheimer’s disease—epidemiological
evidence. Acta Neurol. Scand. Suppl. 185, 50–57
(2006).
118.Kivipelto, M., Solomon, A., Blennow, K.,
Olsson, A. G. & Winblad, B. The new cholesterol
controversy—a little bit of history repeating? Acta
Neurol. Scand. Suppl. 185, 1–2 (2006).
s
w
e
i
v
D
e
E
R
T
e
r
C
u
E
t
R
a
N OR OF
C
O
N
R
U
P
10 | DECEMBER 2009 | volume 5
Acknowledgments
V. Hachinski is funded by the Alzheimer Association,
Award Number IIRG‑08‑91792. P. J. Whitehouse
received support from the National Institute on Aging,
Shigeo and Megumi Takayama, and the Greenwall
Foundation. Barbara Crain, Miia Kivipelto and Michael
Williams made significant suggestions, and we very
much appreciate their critical and thoughtful
comments. We thank Tzipora Sofare, Medical Editor at
the Sandra and Malcolm Berman Brain & Spine
Institute, for her help with the preparation of the
tables and figures.
Supplementary information
Supplementary information is linked to the online
version of the paper at www.nature.com/nrneurol
www.nature.com/nrneurol